Photo-document segmentation method and system

Abstract

The present application provides an improved segmentation method and system for processing digital images that include an imaged document and surrounding image. A plurality of edge detection techniques are used to determine the edges of the imaged document and then segment the imaged document from the surrounding image.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]The present application claims priority under 35 U.S.C. 119(e) to U.S. Provisional Application No. 60 / 927,256 filed May 1, 2007 and U.S. Provisional Application No. 60 / 927,308 filed May 1, 2007, both of which are hereby incorporated by reference.BACKGROUND[0002]1. Technical Field[0003]This application generally relates to a method and system for processing a digital image and, more particularly, for processing a digital image comprising a document image and surrounding image.[0004]2. Description of Related Art[0005]Document management systems are becoming increasingly popular. Such systems ease the burden of storing and handling large databases of documents. To use such systems with paper documents, digital images of the paper documents to be electronically stored must be captured.[0006]There are a variety of potential devices for capturing digital images of paper documents, including digital cameras, scanners, and cellular phones with bu...

AI Technical Summary

Problems solved by technology

It is also more likely that conditions that are less than optimal for existing segmentation techniques—including shadows, poor lighting, indistinct backgrounds, multiple documents in image, and occlusions due to folds, paperclips, tears, etc.—will be encountered when using a camera to capture an image of a document.

Use of a camera can also introduce image distortions, which are generally less prevalent in scanner captured images.

Such distortions may include distortions caused by the angle of the camera relative to the document (“perspective distortion”), the lens used by the camera and its distance from the document, the relative flatness of the document (e.g., whether the document being imaged is situated on a flat or curved surface), and other factors.

As a result, currently known segmentation techniques do not adequately segment the captured image in many situations, particularly when the captured image has been captured using a camera.

Images